Subsections t df Significance (2-tailed) (p-value) Mean Squared Difference Nursing dx as process Nursing dx as product 9.670 98 0.000* 0.442 Nursing dx as process Nursing interventions 9.790 98 0.000* 0.766 Nursing dx as process Nursing-sensitive patient outcomes 9.890 98 0.000* 0.136 Nursing dx as product Nursing Interventions 3.563 98 0.001* 0.449 Nursing dx as product Nursing-sensitive patient outcomes 2.80 98 0.006 0.500 Nursing interventions Nursing-sensitive patient outcomes -0.312 98 0.756 0.640 *Indicated significance as p <0.005 Documentation Forms
Comparison of the quality of documentation in different types of forms was determined by calculating mean scores. These were calculated for the three different types of forms available to nurses to complete their documentation. On average the structured assessment forms available to nursing staff within the Powerchart documentation system were accessed more frequently than the ADHOC and communication forms. The overall mean scores were as follows: structured form x3 = 0.510, communication note x2 = 0.080, and ADHOC forms x1 = 0.045, as
demonstrated in Table 4. Further analysis based on form type was completed by calculating the mean scores for each of the Q-DIO subsections, as seen in Table 5.
31 Table 4: Total Mean Scores for Type of Form
Documentation Form Type Overall Mean Score
Adhoc (x1) 0.045
Communication Note (x2) 0.080
Structured (x3) 0.530
Table 5: Q-DIO Subsection Mean Scores for Type of Form
Subsection of Q- DIO Adhoc Form (Mean Score) (x1) Communication Note (Mean Score) (x2) Structured Form (Mean Score) (x3) Nursing Dx as process 0.140 0.220 0.800 Nursing Dx as product 0.020 0.060 0.440 Nursing interventions 0.020 0.020 0.420 Nursing-sensitive patient outcomes 0.00 0.020 0.380
Structured forms also maintained the highest mean scores when form type for each sub- section was analyzed independently. The mean scores for Nursing Diagnoses as Process were as follows; ADHOC x1 = 0.140, communication note x2 = 0.220, and structured form x3 = 0.800. The mean scores for Nursing Diagnosis as Product were as follows; ADHOC x1 = 0.020, communication note x2 = 0.060, and structured form x3 = 0.440. For Nursing Interventions, the mean scores were ADHOC x1 = 0.020, communication note x2= 0.020, and structured form x3 = 0.420, respectively. Finally, the mean scores for Nursing-Sensitive Patient Outcomes were ADHOC x1 = 0.000, communication note x3 = 0.020, and structured form x3 = 0.380.
Overall, the structured forms, accessible through the basic admissions history and assessment forms provided the highest quality of nursing documentation surrounding
32 psychosocial care. The poorest quality of documentation was demonstrated through the use of ADHOC forms, forms in which nurses had to self-determine use of application.
The mean scores were then subjected to multiple t-tests to determine if there was a
statistical significance in the quality of documentation dependent on the form used. A probability p-value of <0.05 was utilized to determine statistical significance. When comparing structured forms with both the communication notes (p-value 0.000) and ADHOC forms (p-value 0.004), a statistical significance was noted in the quality of documentation. It was also found that a
statistically significant difference in quality also occurred between the use of the ADHOC note compared to the communication note (p-value 0.0337).
Table 6: Comparison of Scores for Type of Form
Comparison of Scores of Different Types of Forms
Type of Form t df Significance (2-tailed) (p- value) Mean Squared Difference Adhoc Communication Note -2.153 98 0.034 0.009 Adhoc Structured -2.988 98 0.004* 0.605 Communication Note Structured -5.89 98 0.000* 0.133 *Indicated significance as p<0.005
Scores pre & post psychosocial form. The Mann-Whitney test was used to test the hypothesis that the distribution of the four Q-DIO subsection elements documented in the patient chart between the pre-psychosocial form and post-psychosocial form period was equal. The difference were tested by (1) the number of different forms where documentation was included (process SUM documented, product SUM documented, intervention SUM documented and
33 outcomes SUM documented) and (2) the number of items documented from the modified Q-DIO (Nsg dx process SUM, Nrsg dx product SUM, Nrsg int. SUM, Nsg outcomes SUM, Total Q-DIo scores).
A statistically significant difference in distribution of documentation within the nursing diagnoses as a process subsection amongst the three different forms was noted. The pre-
psychosocial group had 9.1% of charts documenting nursing diagnoses as a process in all three forms (ADHOC, communication note and structured form), whereas the post-psychosocial form group did not have any charts documenting nursing diagnoses as a process in all three forms (U = 165, p<0.001).
A statistically significant difference was also noted in the distribution of the number of nursing intervention elements documented. Nursing intervention elements comprised of three items: 1) Concrete, clearly named nursing interventions are planned (what will be done, how, how often, who does it?); (2) The nursing interventions effect the aetiology of the nursing diagnosis; and (3) Nursing interventions carried out, are documented (what was done, how, how often, who did it?). The pre-psychosocial form group had 45.5% of charts without any of the three items of nursing intervention documented, whereas the post-psychosocial form had 71.4% of charts without any of the three items documented (U=232.5, p=0.047).
Finally, a statistically significant difference in distribution of total Q-DIO scores was found, with the pre-psychosocial form group having a greater proportion of charts having higher total Q-DIO scores when compared to the post-psychosocial form group (U=211.5, p<0.029). These results are demonstrated in Table 7.
34 Table 7: Mann-Whitney Scores for Form Type and Q-DIO Subsections
Mann - Whitney Scores for Form Type and Q-DIO Subsections
Mann- Whitney U Exact Significance (1- tailed) (p-value) Point Probability Process SUM documented
165.000 0.000* 0.000
Product SUM documented 240.000 0.800 0.027
Interventions SUM documented 228.000 0.058 0.043
Outcomes SUM documented 262.500 0.182 0.077
Nsg dx process SUM 225.500 0.050 0.002
Nsg dx product SUM 232.000 0.051 0.003
Nsg Interventions SUM 232.500 0.047* 0.006
Nsg Outcomes SUM 265.00 0.174 0.011
Total Scores Q-DIO (total of 19 - how
many elements got 1 point) 211.500 0.029* 0.001
*Indicated significance as p<0.005
Seasons and Q-DIO scores. A percentage was calculated to determine the number of charts that fell into each of the specified time periods. In order to fall into the specified time period, the documentation analyzed had to fall within the pre-set time frames of January 1st – March 31st (Winter), April 1st – June 30th (Spring), July 1st – September 30th (Summer), and October 1st – December 31st (Fall). 66% of the charts under analysis contained documentation occurring in the first half of the year 2014 (January 1st – June 30th), and 34% in the second half of the year 2014 (July 1st – December 31st). These time periods were created to further analyze the possible differences in documentation pre and post implementation of the specific
35 Figure 2: Percentage of Charts by Season
The mean scores for each of the Q-DIO subsections were calculated to determine if a difference in the quality of psychosocial nursing documentation occurred at various times during the year. The results of these scores are demonstrated in Table 8. The mean scores of each subsection between the time periods outlined above demonstrated inconsistencies in the quality of documentation. The highest scores, and therefore higher quality of documentation in all sections occurred between January 1st – March 31st, and the lowest scores occurred between October 1st – December 31st. For example, the mean score for Nursing Diagnoses as a Product for January 1st – March 31st was 1.375, compared to the mean scores of 0.000 from October 1st – December 31st. These results are displayed in Table 8.
Table 8: Mean Scores for Q-DIO Subsections Related to Seasons
Seasons Seasons total mean score Nsg. dx as process mean score Nsg. Dx as product mean score Nsg. Int mean score Nsg. Outcomes mean score Jan. 1st - Mar. 31st 5.188 2.438 1.375 0.625 0.813 Apr. 1st - June 30th 5.176 2.353 1.059 1.059 0.706 Jul. 1st - Sept. 30th 2.300 1.400 0.400 0.100 0.400 Oct. 1 -Dec 31st 1.857 1.570 0.000 0.00 0.286
36 Overall the results demonstrated the highest scores occurring in the Nursing Diagnoses as a Process (x1), and the lowest scores occurring in the Nursing Intervention (x3) section. The total mean scores for the Nursing Diagnoses as a Process (x1) compared to Nursing Interventions (x3) were as follows; January 1st to March 31st x
1= 2.438, x3 =0.625; April 1st – June 30th x1 = 2.353, x3 = 1.059; July 1st – September 30th x1 = 1.400, x3 = 0.100 and October 1st – December 31st x1 = 1.570; x3 = 0.000. Therefore, higher quality documentation occurred in the Nursing Diagnoses as a Process section, regardless of the time period within the year. The poorest quality of
documentation was noted in the Nursing Intervention section, regardless of the time period within the year.
Through comparison of the mean scores it was noted that similar levels in the quality of documentation occurred between Nursing Diagnoses as a Product and Nursing-Sensitive Patient outcomes. The comparison pattern based on mean scores demonstrated highest quality of
documentation occurring in the time period of January 1st – March 31st, with a steady decline throughout the subsequent time periods in the year. The lowest quality of documentation in these two sections occurred in the time period of October 1st – December 31st. This comparison demonstrates that the quality of documentation decreased in quality throughout the year. There was no evidence of a quality pattern between any of the other subsections throughout the specified time periods.
37 CHAPTER FOUR
Discussion Q-DIO and Psychosocial Nursing Documentation
The results of this study suggests that the process-based model of the NREM and Q-DIO, can provide detailed information about nurses’ ability to document quality psychosocial care of patients in a palliative care setting. Overall, the results showed consistently poor levels of quality throughout all subsections of the Q-DIO, indicating that nurses had difficulty in applying a process-based approach to documentation of care provided to patients. Nursing diagnoses as a product with a mean score of 2.08 demonstrated the ability of the nurses to document a specific diagnosis. However, mean scores of 0.58, and 0.62 for nursing interventions and nursing- sensitive patient outcomes represent the nurses’ inability to link diagnoses, interventions and outcomes. This finding is consistent with previous evaluation studies examining the
implementation of standardized nursing language which has shown “documentation deficiencies where nursing diagnosis are not coherently linked with nursing interventions and outcomes (Muller-Staub et al., 2006, p.1028)
Quality documentation depends upon the nurse’s ability to make accurate judgments about the patient’s needs using an information feedback loop. Difficulties with this process may be explained by the differences in psychosocial elements. Tangible (objective) information that needs to be documented has been shown to be more prevalent in psychosocial nursing
documentation then intangible (subjective) information throughout nursing documentation (Brooks, 1998; Blair & Smith, 2012; Evans, 2012; Scoates, Fishman & McAdam, 1996). In this study, under the nursing diagnoses as a product subsection, the element with highest quality was “presence of significant others” (mean score 0.80), which is objective information. Nurses in this
38 study were able to visualize the presence of family members at a patient’s bedside making this information tangible The elements with lowest quality were “beliefs and attitudes about life (related to hospitalization)” (mean score 0.28), and “social situation and living environment / circumstances” (mean score 0.16). These two elements are subjective information. Nurses were less able to visualize, or were not aware of the importance of a patient’s social situation or attitudes about life, making this information intangible.
Nurses within this study were able to recognize and identify a key EOL issue
(Gunhardsson, Svensson & Berteroe, 2008; Collier, 2001; Choi et al., 2012). Despite the poor quality in documenting subjective psychosocial issues, the element of anxiety, when documented was consistent and of quality. They were however unable to connect interventions pertaining to anxiety to overall patient outcomes. Although poor quality of psychosocial documentation of responses of patient anxiety does not necessarily indicate that patients had higher levels of anxiety at EOL. Nevertheless, the lack of internal coherence between the three stages of the process suggests that care may be inadequate.
Previous research reports that nurses have been unable to adequately address
psychosocial needs in documentation due to the lack of an accepted nursing language within a streamlined process (Brooks, 1998). The results of this study support these findings since there was a failure to to document on nursing specific interventions provided to patients expressing spiritual needs. This study demonstrates that nurses were limited in their ability to identify spiritual needs, as demonstrated by a mean score of 0.28 for “beliefs and attitudes about life”. Despite being able to identify these needs, they failed to indicate interventions as demonstrated by a mean score of 0.14 for “nursing interventions effects the aetiology of the nursing
39 interventions and outcomes was demonstrated throughout all available types of forms utilized to complete documentation.
Potential Factors Affecting Quality of Documentation
An unexpected result was the significant impact the different types of documentation formats had on quality documentation. The negative implicat of streamlined documentation processes was evident as scores for quality decreased after implementation of a structured psychosocial form within the Powerchart database. Prior to the implementation of this form, the psychosocial documentation under analysis was found in the narrative formats of ADHOC and communication notes. Once this streamlined structured form was implemented, there was a statistical significance in the quality of documentation (U= 211.500, p <0.029).
With fewer patients requiring EOL care (34 admitted between July 1st and December 31st), in combination with a structured streamlined forms, nurses would have more time to conduct quality documentation. The scores were expected to enhance but this was not the case. This significant finding supports previous research studies which identified a limitation with streamlined documentation processes because they do not allow for the nurse to record subjective findings such as “beliefs and attitudes about life” (Scoates, Fishman & McAdam, 1996; Blair & Smith, 2012). Although this form was designed to minimize documenting time, it failed to recognize the importance of providing a process which included narrative
documentation in order to describe psychosocial care. The increased redundancy in
documentation practices, in combination with the lack of standardized nursing language to describe psychosocial care appeared to negatively affect the quality of documentation.
Exploring the effects that educational sessions regarding psychosocial documentation via a structured form has on overall quality is warranted, as previous studies have noted the positive
40 impacts of education on quality documentation. Furthermore, it may be advantageous to
collaborate with an organization to create and implement an educational session regarding the process-based Q-DIO and its implications related to patient outcomes. The educational session could be provided to nursing students and nurses in the palliative care setting to again assess the difference of educational backgrounds on quality documentation.
In summary, this study demonstrated the application of the process-based Nursing Role Effectiveness Model (NREM) and Quality of Documentation of Nursing Diagnoses,
Interventions and Outcomes (Q-DIO) in an area outside of acute medicine. Application of these models, in conjunction with each other, demonstrated that poor quality of psychosocial
documentation may be occurring within a specialized Canadian palliative care setting. Specific issues such as cumbersome charting formats, implementation of structured forms, lack of familiarity with standardized nursing language, and the inability to link nursing diagnoses, interventions and outcomes, may have impacted the overall quality of documentation. This study demonstrated that poor quality in one area or element of documentation (i.e. nursing diagnoses) can negatively influence subsequent areas and illustrates the importance of applying the nursing feedback loop in documentation.
Limitations
Several limitations of this study are noted. The study was completed using a modified Q- DIO, analyzing only specific elements within each subsection of the Q-DIO. A more complex coding format could have been used to allow of a more in depth analysis of the overall Q-DIO scores. Also, a larger sample size or multiple site study would have increased generalizability of the results outside of the palliative care setting. The inclusion of confounding variables such as demographic factors (patient gender, age, diagnoses, co-morbidities), and educational levels of
41 the nurses (registered practical nurse versus registered nurse) could have been assessed to assist in explaining the overall results of the study.
Implications for Further Research
Further research to examine for differences in the type of documentation forms utilized is warranted, as it may lead to development of a streamlined psychosocial documentation process and standardized nursing language. As this study demonstrated it is imperative to evaluate the impact of newly implemented tools (such as structured documentation forms related to
psychosocial care) to determine where issues in documentation are occurring. For example, are the issues occurring at the bedside in terms of the patient actually reporting psychosocial
symptoms, or did the nurse fail to assess for such? As well, future research should include patient specific elements, such as age, gender and diagnoses at time of care, to determine if quality documentation is more prevalent in specific patient situations. More research is needed to develop and test standardized nursing process language and assess its application in a variety of practice settings.
Although this study did not directly assess the influence of nurses’ education levels on the quality of documentation, it did however demonstrate the impact educational levels may have on quality psychosocial documentation. Exploring nursing knowledge and education levels may help to explain the decline in quality after implementation of a designated psychosocial form, as noted in this study. Furthermore, inclusion of nursing designation based on educational levels obtained, may denote that education is directly related to quality of documentation. Inclusion of this type of variable may help to explain any differences in quality of documentation not
attributed to the overall type of form utilized. As well, future research regarding quality psychosocial documentation should include patient specific elements, such as age, gender and
42 diagnoses at time of care, to determine if quality documentation is more prevalent in specific patient situations.
Conclusion
Exploring the current quality of psychosocial nursing documentation at EOL in a Canadian palliative care setting has illustrated the importance of applying a nursing process feedback loop which connects nursing diagnoses, interventions and outcomes. It is apparent from these results that utilizing a process-based framework is a valid technique to assess the quality of nursing documentation which in turn provides insights into the psychosocial care provided to patients. Based on the results of the study, nurses experience challenges in using a process model and therefore further research is needed to identify strategies to promote higher levels of nursing documentation.
Future contributions to nursing research should be centered on establishing Canadian benchmarks in psychosocial nursing documentation at EOL. In order to establish baseline benchmarks, this study recommends the continued analysis of psychosocial documentation to determine overall quality and further evaluation of the process-based NREM and Q-DIO within various Canadian palliative care settings to determine variances in documentation.
Continuing to utilize a process-based model, such as the Q-DIO and nursing feedback loop, to analyze nursing documentation will provide organizations with measureable targets of quality documentation. Prior to implementing newer streamlined documentation models current trends in documentation must be assessed since this study illustrated that it may have a
significant impact. If inadequate documentation is occurring, it can guide future developments in nursing documentation protocols to enhance overall patient outcomes. Once these targets have been established, and alterations to organizational nursing documentation practices are
43 implemented, psychosocial nursing care received by Canadians at end-of-life will be positively impacted.
44 References
Blair, W. & Smith, B. (2012). Nursing documentation: frameworks and barriers. Contemporary
Nurse. 41(2). 160-168.
Bjorvell, C., Wredling, R. & Thorell- Ekstrand, I. (2003). Prerequisites and consequences of nursing documentation in patient records as perceived by group of Registered Nurses.
Journal of Clinical Nursing. 12. 206-214.
Brooks, J. (1998). An analysis of nursing documentation as a reflection of actual nurse work.
Medsurg Nursing. 7(4). 189-198.
Brunoro-Kadash, C., & Kadash, N. (2013). Time to care: A patient-centered quality Improvement strategy. Leadership in Health Services, 26(3), 220-231. doi:http://dx.doi.org/10.1108/LHS-02-2013-
Canadian Medical Protective Association. (2011). Why good documentation matters. Retrieved on January 13th, 2016 from:
http://www.cpso.on.ca/uploadedFiles/members/peerassessment/documentation- CMPA.pdf
Carelock, J. & Innerarity, S. (2001). Critical incidents: effective communication and documentation. Critical Care Nursing Quarterly. 23(4). 59-66.
Choi, M., Lee, J., Kim, S., Kim, D & Kim, H. (2012). Nurses’ knowledge about end-of-life care: where are we?. The Journal of Continuing Education in Nursing. 43(8). 379-384.